Geoestatística de atributos físicos e químicos do solo cultivado com milho irrigado por gotejamento no semiárido pernambucano

Abstract

Soil moisture has a variability in the spatial and temporal domains, leading to uncertainties critical for agricultural water management. The geostatistics allows performing a quantitative description of spatial variability, contributing to the proper management of water and soil. Studies that focus on the spatial variability of physical and chemical properties of soils have great relevance in the literature, mainly as a subsidy for the rational management of water and soil in irrigated perimeters. In order to evaluate the spatial variability of chemical and physical properties of the soil, conducted an experiment in the municipality of Pesqueira, Agreste region of Pernambuco, over a corn crop (Zea mays L.), drip irrigation, using moderately saline water, originating from a well Amazons. The experiment was conducted from 30 March 2015 to 22 June 2015, in an area with spaced mesh of 5 m x 5 m, forming a "grid" 20 m x 25 m, totaling 30 sampling points. The data were evaluated by adopting methods of descriptive statistics, geostatistics and temporal stability using the relative difference and the Spearman correlation. The volumetric Soil moisture was monitored in the layer of 0.00 to 0.20 m using the HFM 2010 sensor. It was found through the semivariogram, spatial dependence of soil moisture with best fit to exponential models and Gaussian. The degrees of spatial dependence showed up moderate and high. Through the contour maps, there are points where the humidity is below the ideal, requiring different management for these areas in order to ensure optimal water supply for the crop. The values of Spearman correlation test found remained high throughout the experiment indicating time dependence throughout the study. Through the relative difference technique it was possible to identify a point that represents the average soil moisture with high reliability, which is recommended for monitoring soil moisture for irrigation management purposes in the area. The model that best fits to the variables clay, sand, electrical conductivity, carbon and organic matter was exponential, while for the Silt the model that best fit was the gaussian. There is a relationship between the spatial distribution of clay with electrical conductivity levels, organic carbon and soil organic matter, and found the highest levels of these attributes in the same regions. Soil texture has significant influences on soil temperature. So that the sampled points that showed higher sand content had higher daily temperature variations in the surface layers. Through the contour maps can identify areas where it is necessary to adopt the practice of application of leaching depths. As for carbon content and organic matter the soil was rated average content of organic matter, requiring the incorporation of organic matter in the soil in regions with low levels of these attributes.